Over the past decade, extracorporeal life support (ECLS) has gained increasing utilization in the management of refractory cardiogenic shock (CS) with a class IIaC recommendation according to the latest European and US guidelines.1 However, recent randomized controlled trials (RCTs) and meta-analyses can be considered inconclusive in demonstrating the benefit of ECLS compared with optimal medical treatment in patients with acute myocardial infarction complicated by CS. Various factors may explain these neutral results but the inherent haemodynamic effect of the ECLS itself probably plays a major role. Due to its aortic retrograde flow principle, peripheral ECLS can lead to left ventricular (LV) overload and distension, increased myocardial workload and pulmonary oedema, compromising cardiac recovery, and diming the patient's prognosis. To optimize the chances of ECLS outcomes with good heart recovery, several left heart decompression (LHD) strategies have been proposed. However, all strategies are invasive, entail risks of bleeding, haemolysis and thromboembolic complications, and the risk–benefit ratio of their use remains unclear.2 Despite limited available data, recent guidelines suggest early initiation of LHD after ECLS initiation (class IIaC) without further defining the preferred unloading technique or its implementation timing after ECLS.1 Left heart decompression techniques aim to unload the left ventricle by increasing the forward flow, achieved either indirectly with the intra-aortic balloon pump (IABP) or directly through microaxial flow pumps, or through left atrial (LA) drainage (atrioseptostomy) or active transapical venting (Figure 1). Active LA venting implies the implementation of an atrioseptostomy in combination with a 21–23 Fr cannula inserted through the interatrial septum and connected in 'Y' to the venous ECLS circuit which remains in place for several days. Its achievement justifies multidisciplinary skills and expertise assuring safe septal defect realization and correct cannula positioning, but also close circuit monitoring during the ECLS run. In this viewpoint around LA venting, we focus on two recent RCTs, EARLY-UNLOAD and EVOLVE-ECLS, conducted in patients with CS undergoing ECLS. In the EARLY-UNLOAD trial (n = 116; single centre), the comparison centred on early unloading through active LA venting within a 12-h window versus a rescue unloading strategy triggered by clinical indications of increased LV afterload.3 The EVOLVE-ECMO trial (n = 60; two centres) examined LHD via LA venting at the time of ECLS insertion (termed the early approach) versus the rescue unloading approach.4 Both trials yielded concordant findings, indicating the absence of significant differences in 30-day mortality rates. It is pertinent to acknowledge that the relatively modest sample sizes in both trials were insufficiently powered to adequately address their primary endpoints. Notably, the sample size calculation for the EARLY-UNLOAD trial was predicated on achieving a 50% reduction in relative mortality risk, assuming an initial unsuitable absolute mortality rate of 25% within the early intervention group. This assumption starkly contrasts with the 50–60% mortality rates reported in the literature for the CS population, underscoring the imperative need for more comprehensive investigations. Moreover, the interpretation of the EARLY-UNLOAD trial is complexified by a high rate of crossover with 41% of patients in the control group that finally underwent LA venting.3 This highlights the significant challenges associated with intention-to-treat analysis, adding a layer of complexity to interpreting the results. The difference in terms of timing of implantation (21.8 h in the control group vs. 1.1 h in the early LHD group) allows us to suggest that early discharge via LA venting is not superior to LHD based on signs of overload, but does not permit to conclude about the efficacy and superiority of LHD by LA venting. Considering the findings of these two trials, two specific points deserve further attention. First, it is essential to determine whether the use of LA venting is an effective unloading technique. Indeed, the utilization of LA venting for LV unloading prompts attention regarding its effectiveness, particularly in scenarios devoid of significant mitral regurgitation. This approach addresses the LV preload, but it disregards the intricate contributions of venous return via Thebesian veins, bronchial veins, and retro-aortic circulation, elements that contribute significantly in case of aortic insufficiency. Furthermore, this indirect method of LV unloading may compromise LV ejection efficiency, hindering adequate opening of the aortic valve and potentially exacerbating stasis, thereby increasing the incidence of LV and aortic thrombus formation. Therefore, the suggested technique may inadvertently overlook crucial facets of cardiac physiology and the intricate dynamics of the circulation. Additionally, the de-escalation process is compromised by the simultaneous withdrawal of LA venting and venoarterial extracorporeal membrane oxygenation (VA-ECMO), not only imposing physiological stress to the myocardium, but also resulting in a persistent septal defect in 17–100% of the survivors, requiring additional closure in up to 50% of cases.5 In opposite to LA venting, the mechanism of both ImpellaTM and IABP offer physiological advantages. The Impella functions as an axial flow pump that actively unloads the left ventricle by propelling blood from the left ventricle into the ascending aorta. This actively assists in reducing myocardial workload and enhancing coronary perfusion. The IABP enhances diastolic coronary perfusion and reduces LV afterload, albeit at a lower extent as compared to the micro-axial flow pumps.6 Secondly, the ideal timing for LHD implementation is a crucial factor for optimizing its effectiveness. Unloading devices can be inserted before, together with, or shortly after the initiation of VA-ECMO. A proactive approach may safeguard the vulnerable ventricle and facilitate myocardial recovery by preventing exposure to elevated afterload, although this should be balanced against the potential risks of increased complications and costs. Based on a recent retrospective multicentre study, Schrage et al.7 showed that early LHD (within the first 2 h after ECLS initiation) was associated with a reduced risk of 30-day mortality (hazard ratio 0.64, 95% confidence interval [CI] 0.46–0.88) and an increased likelihood of successful weaning from mechanical ventilation (odds ratio 2.17, 95% CI 1.19–3.93). It is worth noting that this study specifically examined active LV unloading using the ImpellaTM device. In contrast, despite the remarkably fast median time between randomization and transseptal LA cannulation in the early LHD group, the EARLY-UNLOAD and EVOLVE-ECLS trials failed to confirm the advantages of early LHD. This raises an important question: is the mechanism of unloading/device a contributing factor to the observed outcomes? Although current data suggest that prophylactic unloading after the initiation of VA-ECMO appears beneficial, we can only conclude that further prospective RCTs are needed to establish the optimal timing of LDH initiation. A tailored approach guided by individual haemodynamics and clinical criteria can individualize the unloading strategy and possibly improve outcomes. The only clinical data comparing different unloading strategies rise from meta-analyses pooling retrospective studies.8 Results of these studies highlight the importance of further research in this field comparing different devices and timing of LHD initiation. Ongoing RCTs (UNLOAD-ECMO and REMAP-ECMO trials) are attempting to address the question whether early and/or systematic LHD initiation in CS patients treated with ECLS is beneficial or not. However, challenges in early LHD management and potential complications must be recognized, as they may hinder patient recruitment into RCTs, as seen in the premature termination of the REVERSE trial. Furthermore, additional RCTs comparing different LHD strategies, such as ImpellaTM and IABP, are crucial to evaluate their impact on mortality, morbidity, and myocardial recovery, considering the limited availability and high cost of cardiac grafts and LV assist devices (Table 1). Awaiting randomized data to make firm recommendations on the preferred unloading methods, currently, decisions should be made on a case-by-case pragmatic base, guided by local expertise, and available strategies in the multidisciplinary shock team. Conflict of interest: none declared.
No takes yet. Share an insight, caveat, or question.
Ughetto et al. (2024) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: